enjoying salad since 1978.

Sunday, May 15, 2005

Database Debunkings

"Dispelling persistent prevalent database management fallacies"

With a title like that, how it could be anything less than light, airy, and fun?

Man, I love it when computer scientists get catty. I originally thought this guy might be a huckster using incredibly contrary views to try and beef up his credibility (which works with a surprisingly large number of people) but I think he makes some great points and also offers solutions. I think the attitude is just his way of trying to break past the inertia of these ideas. Hey, it worked on me, I just ordered his book "Practical Issues in Database Management: A Reference for the Thinking Practitioner" from Amazon. It has a forward by C. J. Date, a name which holds some serious weight in the DBMS community.

6 Comments:

Blogger fabian said...

Make sure you apply the errata from the Books page.

Enjoy.

Best regards,

9:01 AM

 
Blogger Greg Stein said...

Man. That site is fugly. I can't imagine how to make it worse.

For example, the "Links" section shows three items. But they aren't links! Instead, you have to click the section title to take you to a page which has... links.

1:10 PM

 
Blogger BillSaysThis said...

I'm surprised Fabian's writings are new to you. I've been reading his stuff back to the early '90s, when I was trying to squeeze relational concepts into my Clipper-ized brain.

7:41 PM

 
Blogger fabian said...

greg,

WOuld you like to maintain that site for me? For free? And make it beautiful too?

Suggest you pay some attention to content, not form.

FP

12:15 AM

 
Blogger jiblet said...

I'd like to read some of their writings, but I can't find anything for free. The dismissal of the entire industry of RDBMS makes me far too skeptical to shell out cash to find out what is wrong with every single 'so-called relational' dbms out there. If the fallacies are so obvious and persistent then there ought to be plenty of free debunkings.

In my extensive coding experience every architecture decision is based on a compromise. They seem to be suggesting that there is an absolutely correct relational setup and schema for every problem and that any performance tradeoffs are the fault of dbms vendors and ultimate data integrity outweighs any performance gains in all cases.

This kind of absolutism rubs me the wrong way as it flies in the face of all my experience with programming in general.

2:21 PM

 
Blogger Steve Jenson said...

jiblet, I would suggest checking out Database in Depth from O'Reilly.

2:26 PM

 

Post a Comment

Links to this post:

Create a Link

<< Home